Quaternary International 400 (2016) 175e179

Contents lists available at ScienceDirect

Quaternary International

journal homepage: www.elsevier.com/locate/quaint

Patterns of evolution in northeast Asia with a particular focus on Salkhit

Damdinsuren Tseveendorj a, Byambaa Gunchinsuren a, Eregzen Gelegdorj a, Seonbok Yi b, * Sang-Hee Lee c, a Institute of Archaeology, Mongolian Academy of Sciences, Ulaanbaatar 51, Mongolia b Department of Archaeology, Seoul National University, Seoul 151-742, Republic of Korea c Department of Anthropology, University of California at Riverside, Riverside, CA 92521-0418, USA article info abstract

Article history: Despite the well published mixture of archaic and modern features in fossil hominins, a presence of Available online 12 February 2016 archaic features is still used as a basis for a claim of an archaic specimen. In this paper, the archaic appearance of a hominin fossil specimen from Salkhit, Mongolia, is examined to ask if Salkhit looks Keywords: archaic because it is an archaic specimen like a classic . The morphology and metrics of the Human evolution Salkhit skullcap was compared with Middle and Late hominin fossils from Zhoukoudian: Paleoanthropology Locality 1 and Upper Cave. Results show that the archaic features that Salkhit shares with the Locality 1 Archaic sample are also shared with the other sample, Upper Cave. On the basis of metrics, Salkhit is interme- Morphology Salkhit diate between the Locality 1 and the Upper Cave specimens. Salkhit is different from the Middle Pleis- Northeast Asia tocene materials in the same way later hominins differ from the Middle Pleistocene sample, in having a broader frontal and thinner supraorbital region. This may reflect encephalization and gracilization, a modernization trend found in many places. Results of this paper are not compatible with the null hy- pothesis that Salkhit is like a member of the Zhoukoudian H. erectus sample. Archaic features may have different explanations: they can be diagnostic features of an archaic species, or regionally predominant features. It is concluded that the latter explains the archaic features of Salkhit. © 2015 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction fossil species alone is not a reliable or accurate biostratigraphic marker of time. It is well documented in paleoanthropology literature that In the absence of chronometric dating, the various dates that archaic features appear in archaic specimens as well as later have been suggested range from early Middle Pleistocene specimens in a mixture of archaic and modern features (Wolpoff, (Tseveendorj et al., 2006; Bae, 2010) to terminal Late Pleistocene 1999). However, it still is a practice often found that a presence of (Kaifu and Fujita, 2012). Coppens et al. (2008) used a multidi- an archaic feature alone provides a basis for a diagnosis of a mensional scaling method to analyze the mixture of archaic and membership in an archaic species. modern features, and concluded that the Salkhit skullcap clusters A skullcap found in Salkhit, Mongolia (48 N, 112 E) was given a with Neandertals, Homo erectus, archaic Homo sapiens, but ruled new genus Mongolanthropus, based on its archaic morphology, with out modern H. sapiens. Coppens et al.'s study focused on the a suggested date of 800,000 years, based on a woolly rhinoceros possible affinity between Salkhit and the Neandertals. If so, it discovered in the vicinity (Tseveendorj et al., 2006). The woolly would extend the Neandertal distribution further east of Okladni- rhinoceros, however, was not found together with the hominin kov Cave, currently the easternmost Neandertal site (Bae, 2010; fossil, and therefore it is uncertain that they are contemporaneous. Derevianko, 2011). Furthermore, woolly rhinoceros has a lengthy tenure of appearance The Salkhit specimen is unlikely to be as old as 800,000 years. It throughout Pleistocene (Boeskorov, 2012) that the presence of the could be of Middle Pleistocene age: hominins left evidence from the late Early Pleistocene in what is now the mainland Asia continent and south Asia (Zhu et al., 2001, 2003, 2004, 2008; Pappu et al., * Corresponding author. 2011). Hominin presence is found in Europe as early as 1 Ma E-mail address: [email protected] (S.-H. Lee). http://dx.doi.org/10.1016/j.quaint.2015.08.074 1040-6182/© 2015 Elsevier Ltd and INQUA. All rights reserved. 176 D. Tseveendorj et al. / Quaternary International 400 (2016) 175e179

(Dennell, 2003). Given the long history of hominin occupation in difference in the anterior projection between the medial and the Asia and Europe, this paper considers the possibility that Salkhit is lateral portions. The lateral portion of the supraorbital region turns Middle Pleistocene in age. into a slight knob at the lateral ends in both Salkhit and Locality 1 sample; however, the thickening at the lateral ends of the supra- 2. Methods and materials orbitals is not comparable to that seen in the Zhoukoudian sample which has a knobbed look. In Salkhit the lateral torus is continuous, The Salkhit skullcap is compared with the only hominin sample but more gracile than the medial portion; in Locality 1, the lateral of Middle and Late Pleistocene specimens from the geographic vi- portion is thicker and smoother, without any interruption in the cinity: Zhoukoudian Locality 1 and Upper Cave. Other hominin surface of the torus; the supraorbital torus reaches maximum fossils are isolated individual finds. Because only the skullcap is thickness toward the lateral end to form a knob. preserved, comparisons were limited, based on morphology and In Salkhit, there is a weak supraorbital sulcus that does not metrics. Observation and measurements for the Salkhit skullcap continue across the midline. In contrast, in the Locality 1 sample, were taken on the original specimen as well as on a cast. Photo- the sulcus is an actual gutter that is dipping (as in XII) or graphs are from the original specimen (Fig. 1). Comparison with shallow and broad (as in Skull V), and is most pronounced where other specimens was done with casts only. the supraorbital torus is the thickest. In Salkhit and Locality 1 crania, the supraorbitals are different 3. Results and discussion morphologically between the medial and the lateral halves. The contrast between the medial and the lateral portions of the su- 3.1. Morphological comparison praorbital region of Salkhit takes a different pattern in the Locality 1 sample. In the Locality 1 sample, the medial and lateral portions do At first glance, Salkhit shares similarity in several morphological not have an incisura. The medial portion is laterally marked by a traits with Locality 1. Salkhit has a weak sagittal keel on the frontal supraorbital process (most prominent in Skull X). There is an inci- between glabella and bregma, but there is no prebregmatic sura in Salkhit, which is noted as a modern human feature by eminence. However, the sagittal keel in Salkhit is not as prominent Weidenreich (1943 p. 29). A supraorbital foramen is absent in the as in the Zhoukoudian specimens, most prominently expressed in Locality 1 sample, while a double-notch is in the location of the Zhoukoudian XII, and comparably weak in Upper Cave 101. supraorbital process in Skull XII. The supraorbital region of the Salkhit skullcap plays a major role When Salkhit is compared with the three specimens in Upper in giving an archaic appearance of the specimen (Fig. 2). Seen from Cave, two specimens, UC 102 and UC 103, have a superciliary arch the front, the torus forms an M or a seagull-shape, contributed by and do not have a supraorbital torus. UC 101 shows a surprising the pinching of the glabellar torus. The glabellar torus rises above similarity with Salkhit in many aspects. In both Salkhit and UC 101, the nasal bridge and is thinner in supero-inferior height than the the supraorbital bar continues throughout the supraorbtial region; supraorbital torus in Salkhit and in some Locality 1 crania (espe- the medial portions are thicker than the lateral portions; the su- cially in Skull XII (L3)) and Upper Cave (UC 101), but not in others praorbital torus thickens at the lateral end, resulting in a slightly (such as Skull V). knobbed look. There is a weak but discernible groove that separates The medial portion of the supraorbitals has greater supero- the torus into two components, medial and lateral. In Salkhit, the inferior thickness than the lateral portion, but there is no medial and the lateral portions of the left supraorbital are

Fig. 1. Salkhit skullcap: (a) superior view; (b) lateral view; (c) anterior view. D. Tseveendorj et al. / Quaternary International 400 (2016) 175e179 177

Fig. 2. Anterior views showing variation in supraorbital regions of Salkhit and Locality 1 specimens. The specimens are not in the same orientation. demarcated by a faint groove, which is weaker than is observed in 3.2. Metric comparison UC 101; it is almost an uninterrupted continuation of a torus on the right side, but it cannot be ascertained due to a possible distortion In metric measurements, some variables show Salkhit to be in- from the healed wound. between Locality 1 and Upper Cave (Fig. 3), while other variables In the parietals, there is no keel on the sagittal suture of Salkhit, are non-informative. Due to the small sample size, statistical tests in contrast to the Locality 1 specimens. In both Salkhit and UC 101, were not performed. there is a raised ridge that runs parallel to the sagittal suture, and In frontal breadth, Salkhit at 98.06 mm is greater than the range the ridges follow the suture, diverging as they swerve posterior and of the Locality 1 sample (X, XI, XII, V), from 85.7 to 89 mm inferior. This feature is also seen in Locality 1 Skull XII. (mean ¼ 87.5 mm), and smaller than the range of the Upper Cave In Salkhit, the nasofrontal and frontomaxillar sutures form a sample (from 102 to 109.3 mm). Frontal length is not informative, continuous, horizontal course, similar to the case in Locality 1 Skull as the range in the Locality 1 sample encompasses Salkhit and XII. In Skull XII, however, nasal bones are broad, with no difference Upper Cave specimens (Table 1). However, the frontal curvature between the upper breadth and the middle breadth, while in Sal- index (¼(arc/chord)*100) of Salkhit (109.12) is larger than the khit, the nasals have a pinched shape. Salkhit has a low nasal angle Zhoukoudian values. (height of the saddle made from two nasals); Upper Cave speci- mens have a higher nasal angle than Salkhit. Table 1 Frontal metric data (in mm.). Osteometric points ft ¼ frontotemporale;gl¼ glabella; br ¼ bregma. M9 refers to the osteometric point number 9 using the Martin numbering system (Martin and Saller, 1957).

Ft-ft (M9) Gl-br arc Gl-br chord Curvature

Salkhit 98.06 117 107.22 109.12 ZKD10 89 120.5 114.7 105.06 ZKD11 85.7 114 105.4 108.16 ZKD12 87.9 110 107 102.80 ZKD5 87.4 122 116.1 105.08 ZKD2 84 115 111.5 103.14 UC 101 109.30 117.50 108.80 108.00 UC 102 104.00 119.50 111.00 107.66 UC 103 102.00 113.00 104.00 108.65

4. Discussion

The case of Salkhit is another example where an archaic mem- bership is suggested based on an archaic or a robust appearance. The results of this paper provide support that archaic features do not necessarily mark the specimen to be from an archaic time. An Fig. 3. Scatterplot of frontal length (chord) versus frontal breadth. Salkhit is between the Zhoukoudian Locality 1 specimens and the Zhoukoudian Upper Cave specimens in archaic morphology may be an indication of an archaic ancestry, frontal breadth. rather than an archaic time period. 178 D. Tseveendorj et al. / Quaternary International 400 (2016) 175e179

The similarities that Salkhit share Zhoukoudian Locality 1 are Acknowledgements the same ones that it shares with Zhoukoudian Upper Cave. Such features, for example, sagittal keel, are examples of regionally We thank Robin Dennell and Gao Xing for the invitation to persistent features (Rosenberg and Wu, 2013), found in Dali and participate in this special issue. We give heartfelt gratitude to Ziyang (Wolpoff, 1999) in addition to Salkhit and Zhoukoudian. Yves Coppens (Centre National de Recherche Scientifique) and That a specimen such as Salkhit shows such a trait is not supportive Milford Wolpoff (Univeristy of Michigan) for their infinite gen- of its antiquity; rather, it is compatible with its regional origin. erosity. Robin Dennell and two reviewers provided generous help Despite the archaic impression from the supraorbital region, Salkhit in strengthening this paper. This study was supported by Na- is otherwise more gracile than the Zhoukoudian H. erectus sample; tional Geographic Society (W35-08), National Science Founda- Salkhit's cranial thickness, measured at bregma (6 mm) is thinner tion (0803410), Seoul National University, and the University of than the Locality 1 range (7e10.6 mm). California at Riverside. A similar example of archaic features not being a marker for an archaic specimen is also found in the case of Ceprano, of which the initial date of 1 myr was compatible with its archaic features References (Manzi, 2004); however, its date is now considered to be half as old, 460 kyr (Muttoni et al., 2009), which puts the archaic features of Anton, S.C., 2003. Natural history of Homo erectus. American Journal of Phys- Ceprano in a new light (Dennell et al., 2011). Ceprano is a member ical Anthropology, Supplement: Yearbook of Physical Anthropology 122, 126e170. of a population that exhibits archaic features due to ancestry, and Bae, C.J., 2010. The late Middle Pleistocene hominin fossil record of eastern Asia: not because of an archaic date. Synthesis and review. American Journal of Physical Anthropology 143, 75e93. The morphology of Salkhit supraorbital region can be Boeskorov, G.G., 2012. Some specific morphological and ecological features of the described as intermediate between a supraorbital torus and a fossil woolly rhinoceros (Coelodonta antiquitatis Blumenbach 1799). Biology Bulletin 39, 692e707. superciliary arch. Whether it is a supraorbital torus or superciliary Coppens, Y., Tseveendorj, D., Demeter, F., Turbat, T., Giscard, P.-H., 2008. Discovery of arch is somewhat irrelevant in the investigation into affinity: the an archaic Homo sapiens skullcap in Northeast Mongolia. Comptes Rendus e mere presence of torus does not define Salkhit as an archaic Palevol 7, 51 60. Crooson, D.L., 2010. Vieus d’environ 20,000 ans l’homme de Salkhit a parle. National specimen. A supraorbital torus is not exceptional or unusual Geographic 2010, 49e57. among extant, recent , such as Mladec(Frayer et al., Dennell, R., 2003. Dispersal and colonisation, long and short chronologies: how 2006), Willandra Lake Hominid 50 and several from the Coo- continuous is the Early Pleistocene record for hominids outside East Africa? Journal of Human Evolution 45, 421e440. bool Creek sample (Thorne and Wolpoff, 1981) being only a few of Dennell, R.W., Martinon-Torres, M., Bermúdez de Castro, J.M., 2011. Hominin vari- many examples. M-shaped supraorbitals are also found in speci- ability, climatic instability and population demography in Middle Pleistocene mens from Middle and Late Pleistocene, and also in Australians as Europe. Quaternary Science Reviews 30, 1511e1524. Derevianko, A.P., 2011. The origin of anatomically modern humans and their recent as WLH 50 (Durband and Westaway, 2013). Considering behavior in africa and eurasia. Archaeology, Ethnology and Anthropology of the recent discovery of Denisovan genetic signal in southeast Asia Eurasia 39, 2e31. while almost absent in northeast Asia (Reich et al., 2010, 2011; Durband, A.C., Westaway, M.C., 2013. Perspectives on the origins of modern Aus- tralians. In: Smith, F.H., Ahern, J.C.M. (Eds.), Origins of Modern Humans: Biology Skoglund and Jakobsson, 2011), the shared similarity between Reconsidered, second ed. Wiley, New York, pp. 123e150. Salkhit and the recent Australians points toward a future Frayer, D.W., Jelínek, J., Oliva, M., Wolpoff, M.H., 2006. Aurignacian male crania, research. jaws, and teeth from the Mladec Caves, Moravia, Czech Republic. In: Teschler- Salkhit is different from the Zhoukoudian H. erectus sample in Nicola, M. (Ed.), Early Modern Humans at the Moravian Gate: the Mladec Caves and Their Remains. Springer, Wien, pp. 185e272. the same ways that modern humans are different from H. erectus, Kaifu, Y., Fujita, M., 2012. Fossil record of early modern humans in East Asia. Qua- namely in the expansion of brain size and gracilization. Metric ternary International 248, 2e11. variables suggest a trend with a direction. While frontal length Lee, S.-H., Wolpoff, M.H., 2002. Pattern of brain size increase in Pleistocene Homo (abstract). Journal of Human Evolution 42, A19eA20. measurements do not change, as all specimens show values within Lordkipanidze, D., Ponce de Leon, M.S., Margvelashvili, A., Rak, Y., Rightmire, G.P., the range defined by the Zhoukoudian Locality 1 sample, there is a Vekua, A., Zollikofer, C.P.E., 2013. A complete skull from Dmanisi, Georgia, and e distinct pattern in the frontal breadth measurements. Salkhit is in- the evolutionary biology of early Homo. Science 342, 326 331. Manzi, G., 2004. Human evolution at the Matuyama-Brunhes boundary. Evolu- between the Zhoukoudian Locality 1 and the Upper Cave speci- tionary Anthropology 13, 11e24. mens, suggesting an expansion in frontal breadth. This encephali- Martin, R., Saller, K., 1957. Lehrbuch der Anthropologie. Gustav Fischer Verlag, zation trend reflects what is found throughout the Pleistocene (Lee Stuttgart. Muttoni, G., Scardia, G., Kent, D.V., Swisher, C.C., Manzi, G., 2009. Pleistocene and Wolpoff, 2002). A corollary to that observation is that Salkhit magnetochronology of early hominin sites at Ceprano and Fontana Ranuccio, has less postorbital constriction than the Zhoukoudian sample. The Italy. Earth and Planetary Science Letters 286, 255e268. timing of cranial increase in frontal breadth is compatible with Pappu, S., Gunnell, Y., Akhilesh, K., Braucher, R.G., Taieb, M., Demory, F.ß., Thouveny, N., 2011. Early pleistocene presence of Acheulian hominins in south gracilization and encephalization, the two trends that are observed India. Science 331, 1596e1599. globally throughout Late Pleistocene human evolution (Wolpoff, Reich, D., Green, R.E., Kircher, M., Krause, J., Patterson, N., Durand, E.Y., Viola, B., 1999). Briggs, A.W., Stenzel, U., Johnson, P.L.F., Maricic, T., Good, J.M., Marques- The morphological configuration of Salkhit is quite compatible Bonet, T., Alkan, C., Fu, Q., Mallick, S., Li, H., Meyer, M., Eichler, E.E., Stoneking, M., Richards, M., Talamo, S., Shunkov, M.V., Derevianko, A.P., with the date mentioned in the magazine interview (Crooson, Hublin, J.-J., Kelso, J., Slatkin, M., Pa€abo,€ S., 2010. Genetic history of an archaic 2010), 27,000 years old. Although Salkhit may not look like a hominin group from Denisova Cave in Siberia. Nature 468, 1053e1060. fi member of Zhoukoudian H. erectus, it is still possible that Salkhit is Reich, D., Patterson, N., Kircher, M., Del n, F., Nandineni, M.R., Pugach, I., Ko, A.M.-S., Ko, Y.-C., Jinam, T.A., Phipps, M.E., Saitou, N., Wollstein, A., Kayser, M., Pa€abo,€ S., a Middle Pleistocene specimen. This has particular relevance as Stoneking, M., 2011. Denisova admixture and the first modern human dispersals questions arise whether Zhoukoudian is representative of Middle into Southeast Asia and Oceania. The American Journal of Human Genetics 89, e Pleistocene Asian H. erectus: Anton argues that Zhoukoudian is an 516 528. Rosenberg, K.R., Wu, X., 2013. A river runs through it: modern human origins in East exception rather than a rule for Asian H. erectus (Anton, 2003). Asia. In: Smith, F.H., Ahern, J.C.M. (Eds.), Origins of Modern Humans: Biology Considering that more research is showing early Homo to be more Reconsidered, second ed. Wiley, New York, pp. 89e121. variable that was thought before (Van Arsdale and Wolpoff, 2012; Skoglund, P., Jakobsson, M., 2011. Archaic human ancestry in East Asia. Proceedings of the National Academy of Sciences 108, 18301e18306. Lordkipanidze et al., 2013), a wider range of specimens are neces- Thorne, A.G., Wolpoff, M.H., 1981. Regional continuity in Australasian Pleistocene sary to provide a comparative perspective. hominid evolution. American Journal of Physical Anthropology 55, 337e349. D. Tseveendorj et al. / Quaternary International 400 (2016) 175e179 179

Tseveendorj, D., Batbold, N., Amgalantogs, T., 2006. Mongolanthropus was discov- Zhu, R.X., Hoffman, K.A., Potts, R., Deng, C.L., Pan, Y.X., Guo, B., Shi, C.D., Guo, Z.T., ered in Mongolia. Studia Archeologica Instituti Archaeologici Academiae Sci- Yuan, B.Y., Hou, Y.M., Huang, W.W., 2001. Earliest presence of humans in entiarum Mongolicae 23, 5e10 (in Mongolian). northeast Asia. Nature 413, 413e417. Van Arsdale, A.P., Wolpoff, M.H., 2012. A single lineage in early Pleistocene Homo: Zhu, R.X., Potts, R., Pan, Y.X., Yao, H.T., L¸ , L.Q., Zhao, X., Gao, X., Chen, L.W., Gao, F., size variation continuity in early Pleistocene Homo crania from East Africa and Deng, C.L., 2008. Early evidence of the genus Homo in East Asia. Journal of Georgia. Evolution 67, 841e850. Human Evolution 55, 1075e1085. Weidenreich, F., 1943. The Skull of Sinanthropus Pekinensis: a Comparative Study of a Zhu, R.X., Potts, R., Xie, F., Hoffman, K.A., Deng, C.L., Shi, C.D., Pan, Y.X., Wang, H.Q., Primitive Hominid Skull. Palaeontologia Sinica. New Series D 10. Shi, R.P., Wang, Y.C., Shi, G.H., Wu, N.Q., 2004. New evidence on the earliest Wolpoff, M.H., 1999. Paleoanthropology, second ed. McGraw-Hill, New York. human presence at high northern latitudes in northeast Asia. Nature 431, Zhu, R., An, Z., Potts, R., Hoffman, K.A., 2003. Magnetostratigraphic dating of early 559e562. humans in China. Earth-Science Reviews 61, 341e359.